Experience

Education and Experiences:

2003-2007

BS in Statistics

Nankai University, China

Highlights

1. Scholarships for Academic Distinction, Nankai University, China, 2003-2007.

2. National Mathematical Modeling Contest Winner, Department of Mathematics, Nankai University, China, 2007.

2007-2010

PhD in Statistics

Duke University, US

Title: Nonparametric Bayes Models for High-Dimensional and Sparse Data

Adviser : Prof David Dunson

Description: My Ph.D. thesis studies the applications and robustness of Nonparametric Bayesian paradigms in characterizing large complex data structures from a broad variety of problems.

Highlights

1. Fellowship, Duke University, 2007-2008.

2. Research Assistantship, Duke University, 2008-2010.

3. NSF travel award for participating in International Society for Bayesian Analysis (ISBA) World Meetings, Benidorm, Spain, 2010.

4. Best Poster Award, Frontier of Statistical Decision Making and Bayesian Analysis, 2010.

5. Classification Society (CS) Distinguished Dissertation Award, 2011.

2011-2014

Research Staff Member

IBM T.J. Watson Research Center, Yorktown Heights, NY

Highlights

Academic:

1. Published around 20 Journal/Conference papers.

2. Invited to give around 20 academic talks in different conferences and universities.

3. Held 9 filed US patents.

4. Session organizers/chairs for major statistical and machine learning conferences.

5. Associate Editor, Applied Stochastic Models in Business and Industry .

Business:

1. Statistical Tracking and Analysis for Worldwide IBM Revenue Forecasting which cover thousands of economic regions and hundreds of line of business.

2. Talents Allocation Optimization to Improve IBM Worldwide Sales affecting hundreds of thousands of IBM people and projects.

3. Dynamic Automatic Pricing Optimization System that have been implemented by two Fortune Global 500 companies.

2015-2016/09

Principal Data Scientist on Computational Advertisement

Yahoo! Inc, Sunnyvale, CA

Highlights

Focus on developing flexible (non)Bayesian modeling and algorithm that can be scalable to extremely large data to be deployed on Yahoo! advertising platform, Brightroll, for real time prediction and bidding. Representative works include:

1. CTR Prediction: Multidimensional Dynamic Hierarchical Bayesian Framework (KDD 2015 & ASMBI 2016).

2. Text Analyses: Bayesian Reinforcement Learning for Word Representations (Yahoo! Tech Pulse 2015).

3. Recommendations: User Action Prediction Using Local Graph Algorithms on Massive Data (KDD 2016).

4. CVR Prediction: Dynamic Bayesian Transfer Learning of Global and Local Features (KDD 2016).

5. Meta Analyses for Dynamic Contextual Multi Arm Bandits in Display Advertisement (ICDM 2016).

Editor, Applied Stochastic Models in Business and Industry , Bayesian Statistics and Machine Learning in Business , 2016. We invited top researchers from both academia and industry (Amazon, Google Research, IBM T.J. Watson Research Center, Linkedin, Microsoft Research, Yahoo! etc) to contribute papers and discussions.

2016/09-Now

Senior Staff Data Scientist and Director

Alibaba Group

Highlights

Focus on inductive reasoning with graph learning and their applications to recommender systems.

1. Winner of the Super AI Leader of World Artificial Intelligence Conference 2019 (SAIL).

2. Winner of the Leading Innovation Team of Hangzhou City 2019.

3. More than 25 top conference papers in the related field.